fix-issue
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from GitHub issues and comments via
gh issue view --comments. This content is used to guide the agent's actions, including code modification and testing. - Ingestion points: SKILL.md (Step 1).
- Boundary markers: None; instructions explicitly direct the agent to 'read ALL comments'.
- Capability inventory: Shell command execution (gh, git, terraform, rg), local script execution (scripts/.sh, scripts/.py), and git push operations.
- Sanitization: None provided.
- [COMMAND_EXECUTION]: The skill interpolates user-provided arguments (issue_number) and issue-derived strings (logins, keywords) into shell commands. This pattern is vulnerable to command injection if the agent does not strictly validate the inputs before execution.
- [DYNAMIC_EXECUTION]: The skill executes various project-internal scripts such as
scripts/v2_to_v3_migration_assistant.pyandscripts/cleanup.shto perform migration and teardown tasks.
Audit Metadata